SparkFun Forums

[Kahuna]

I'm running a pro-mini 5v Arduino with a 12v power source. I assumed that it would get warm when being powered by 12v, but it is very hot. Too hot to touch in fact and I got 150c measuring with an IR thermometer.

Is this “normal”? If so can this thing really run this hot all the time with no heatsink? Seems to me like this isn't right.

Thanks!

[tecoist]

Which part is hot, the AVR or the Micrel linear regulator (or both)? Neither should ever run that hot, of course; as you can see in the datasheets, they both specify 125 degrees Celsius as their maximum (and at least in the case of the Micrel regulator, that's the junction temperature inside the package, so the surface temperature has to be somewhat cooler).

What loads are you trying to drive with the pro mini board? At 12V input, you're limited to about 60mA total for the microcontroller and all its loads, because most of your input power (7/12, to be precise) gets burned up in the regulator, and more than 60mA would overheat it (and, in theory, activate its thermal shutdown).

[Kahuna]

Thanks for the quick reply. Its the AVR thats baking. I'm actually not drawing anything on it yet, this is just hooking up power to the board.

[tecoist]

Something's very wrong, then. I hope you have the 12V connected to the Micrel input (RAW on the schematic) and not to Vcc.

The pro mini is pretty simple. If you connect 12V to the RAW input, 7/12 of the power goes into the Micrel regulator, and 5/12 goes into the AVR (and any loads it drives). So logically, if the AVR is burning enough power to get to 150 degrees, the regulator should long since have gone into thermal shutdown. On the other hand, if you accidentally connected 12V to Vcc, the AVR would pretty much melt (much as you describe), probably irreversibly.

[Kahuna]

Good call, that's exactly what I did. I'm guessing this thing is toast. I'll try later today to see if it shows any signs of life.

Thanks for getting me set straight.

[tecoist]

Good luck! Sorry I was right, needless to say, though I suppose anything that got the microcontroller to 150 degrees would be equally unfortunate.